1 // std::ctype implementation details, GNU version -*- C++ -*- 2 3 // Copyright (C) 2001-2018 Free Software Foundation, Inc. 4 // 5 // This file is part of the GNU ISO C++ Library. This library is free 6 // software; you can redistribute it and/or modify it under the 7 // terms of the GNU General Public License as published by the 8 // Free Software Foundation; either version 3, or (at your option) 9 // any later version. 10 11 // This library is distributed in the hope that it will be useful, 12 // but WITHOUT ANY WARRANTY; without even the implied warranty of 13 // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 14 // GNU General Public License for more details. 15 16 // Under Section 7 of GPL version 3, you are granted additional 17 // permissions described in the GCC Runtime Library Exception, version 18 // 3.1, as published by the Free Software Foundation. 19 20 // You should have received a copy of the GNU General Public License and 21 // a copy of the GCC Runtime Library Exception along with this program; 22 // see the files COPYING3 and COPYING.RUNTIME respectively. If not, see 23 // <http://www.gnu.org/licenses/>. 24 25 // 26 // ISO C++ 14882: 22.2.1.1.2 ctype virtual functions. 27 // 28 29 // Written by Benjamin Kosnik <bkoz@redhat.com> 30 31 #include <locale> 32 #include <cstdio> 33 #include <bits/c++locale_internal.h> 34 35 namespace std _GLIBCXX_VISIBILITY(default) 36 { 37 _GLIBCXX_BEGIN_NAMESPACE_VERSION 38 39 // NB: The other ctype<char> specializations are in src/locale.cc and 40 // various /config/os/* files. 41 ctype_byname<char>::ctype_byname(const char* __s, size_t __refs) 42 : ctype<char>(0, false, __refs) 43 { 44 if (std::strcmp(__s, "C") != 0 && std::strcmp(__s, "POSIX") != 0) 45 { 46 this->_S_destroy_c_locale(this->_M_c_locale_ctype); 47 this->_S_create_c_locale(this->_M_c_locale_ctype, __s); 48 this->_M_toupper = this->_M_c_locale_ctype->__ctype_toupper; 49 this->_M_tolower = this->_M_c_locale_ctype->__ctype_tolower; 50 this->_M_table = this->_M_c_locale_ctype->__ctype_b; 51 } 52 } 53 54 ctype_byname<char>::~ctype_byname() 55 { } 56 57 #ifdef _GLIBCXX_USE_WCHAR_T 58 ctype<wchar_t>::__wmask_type 59 ctype<wchar_t>::_M_convert_to_wmask(const mask __m) const throw() 60 { 61 __wmask_type __ret; 62 switch (__m) 63 { 64 case space: 65 __ret = __wctype_l("space", _M_c_locale_ctype); 66 break; 67 case print: 68 __ret = __wctype_l("print", _M_c_locale_ctype); 69 break; 70 case cntrl: 71 __ret = __wctype_l("cntrl", _M_c_locale_ctype); 72 break; 73 case upper: 74 __ret = __wctype_l("upper", _M_c_locale_ctype); 75 break; 76 case lower: 77 __ret = __wctype_l("lower", _M_c_locale_ctype); 78 break; 79 case alpha: 80 __ret = __wctype_l("alpha", _M_c_locale_ctype); 81 break; 82 case digit: 83 __ret = __wctype_l("digit", _M_c_locale_ctype); 84 break; 85 case punct: 86 __ret = __wctype_l("punct", _M_c_locale_ctype); 87 break; 88 case xdigit: 89 __ret = __wctype_l("xdigit", _M_c_locale_ctype); 90 break; 91 case alnum: 92 __ret = __wctype_l("alnum", _M_c_locale_ctype); 93 break; 94 case graph: 95 __ret = __wctype_l("graph", _M_c_locale_ctype); 96 break; 97 case blank: 98 __ret = __wctype_l("blank", _M_c_locale_ctype); 99 break; 100 default: 101 __ret = __wmask_type(); 102 } 103 return __ret; 104 } 105 106 wchar_t 107 ctype<wchar_t>::do_toupper(wchar_t __c) const 108 { return __towupper_l(__c, _M_c_locale_ctype); } 109 110 const wchar_t* 111 ctype<wchar_t>::do_toupper(wchar_t* __lo, const wchar_t* __hi) const 112 { 113 while (__lo < __hi) 114 { 115 *__lo = __towupper_l(*__lo, _M_c_locale_ctype); 116 ++__lo; 117 } 118 return __hi; 119 } 120 121 wchar_t 122 ctype<wchar_t>::do_tolower(wchar_t __c) const 123 { return __towlower_l(__c, _M_c_locale_ctype); } 124 125 const wchar_t* 126 ctype<wchar_t>::do_tolower(wchar_t* __lo, const wchar_t* __hi) const 127 { 128 while (__lo < __hi) 129 { 130 *__lo = __towlower_l(*__lo, _M_c_locale_ctype); 131 ++__lo; 132 } 133 return __hi; 134 } 135 136 bool 137 ctype<wchar_t>:: 138 do_is(mask __m, wchar_t __c) const 139 { 140 // The case of __m == ctype_base::space is particularly important, 141 // due to its use in many istream functions. Therefore we deal with 142 // it first, exploiting the knowledge that on GNU systems _M_bit[5] 143 // is the mask corresponding to ctype_base::space. NB: an encoding 144 // change would not affect correctness! 145 bool __ret = false; 146 if (__m == _M_bit[5]) 147 __ret = __iswctype_l(__c, _M_wmask[5], _M_c_locale_ctype); 148 else 149 { 150 // Highest bitmask in ctype_base == 11 151 const size_t __bitmasksize = 11; 152 for (size_t __bitcur = 0; __bitcur <= __bitmasksize; ++__bitcur) 153 if (__m & _M_bit[__bitcur]) 154 { 155 if (__iswctype_l(__c, _M_wmask[__bitcur], _M_c_locale_ctype)) 156 { 157 __ret = true; 158 break; 159 } 160 else if (__m == _M_bit[__bitcur]) 161 break; 162 } 163 } 164 return __ret; 165 } 166 167 const wchar_t* 168 ctype<wchar_t>:: 169 do_is(const wchar_t* __lo, const wchar_t* __hi, mask* __vec) const 170 { 171 for (; __lo < __hi; ++__vec, ++__lo) 172 { 173 // Highest bitmask in ctype_base == 11 174 const size_t __bitmasksize = 11; 175 mask __m = 0; 176 for (size_t __bitcur = 0; __bitcur <= __bitmasksize; ++__bitcur) 177 if (__iswctype_l(*__lo, _M_wmask[__bitcur], _M_c_locale_ctype)) 178 __m |= _M_bit[__bitcur]; 179 *__vec = __m; 180 } 181 return __hi; 182 } 183 184 const wchar_t* 185 ctype<wchar_t>:: 186 do_scan_is(mask __m, const wchar_t* __lo, const wchar_t* __hi) const 187 { 188 while (__lo < __hi && !this->do_is(__m, *__lo)) 189 ++__lo; 190 return __lo; 191 } 192 193 const wchar_t* 194 ctype<wchar_t>:: 195 do_scan_not(mask __m, const char_type* __lo, const char_type* __hi) const 196 { 197 while (__lo < __hi && this->do_is(__m, *__lo) != 0) 198 ++__lo; 199 return __lo; 200 } 201 202 wchar_t 203 ctype<wchar_t>:: 204 do_widen(char __c) const 205 { return _M_widen[static_cast<unsigned char>(__c)]; } 206 207 const char* 208 ctype<wchar_t>:: 209 do_widen(const char* __lo, const char* __hi, wchar_t* __dest) const 210 { 211 while (__lo < __hi) 212 { 213 *__dest = _M_widen[static_cast<unsigned char>(*__lo)]; 214 ++__lo; 215 ++__dest; 216 } 217 return __hi; 218 } 219 220 char 221 ctype<wchar_t>:: 222 do_narrow(wchar_t __wc, char __dfault) const 223 { 224 if (__wc >= 0 && __wc < 128 && _M_narrow_ok) 225 return _M_narrow[__wc]; 226 #if __GLIBC__ > 2 || (__GLIBC__ == 2 && __GLIBC_MINOR__ > 2) 227 __c_locale __old = __uselocale(_M_c_locale_ctype); 228 #endif 229 const int __c = wctob(__wc); 230 #if __GLIBC__ > 2 || (__GLIBC__ == 2 && __GLIBC_MINOR__ > 2) 231 __uselocale(__old); 232 #endif 233 return (__c == EOF ? __dfault : static_cast<char>(__c)); 234 } 235 236 const wchar_t* 237 ctype<wchar_t>:: 238 do_narrow(const wchar_t* __lo, const wchar_t* __hi, char __dfault, 239 char* __dest) const 240 { 241 #if __GLIBC__ > 2 || (__GLIBC__ == 2 && __GLIBC_MINOR__ > 2) 242 __c_locale __old = __uselocale(_M_c_locale_ctype); 243 #endif 244 if (_M_narrow_ok) 245 while (__lo < __hi) 246 { 247 if (*__lo >= 0 && *__lo < 128) 248 *__dest = _M_narrow[*__lo]; 249 else 250 { 251 const int __c = wctob(*__lo); 252 *__dest = (__c == EOF ? __dfault : static_cast<char>(__c)); 253 } 254 ++__lo; 255 ++__dest; 256 } 257 else 258 while (__lo < __hi) 259 { 260 const int __c = wctob(*__lo); 261 *__dest = (__c == EOF ? __dfault : static_cast<char>(__c)); 262 ++__lo; 263 ++__dest; 264 } 265 #if __GLIBC__ > 2 || (__GLIBC__ == 2 && __GLIBC_MINOR__ > 2) 266 __uselocale(__old); 267 #endif 268 return __hi; 269 } 270 271 void 272 ctype<wchar_t>::_M_initialize_ctype() throw() 273 { 274 #if __GLIBC__ > 2 || (__GLIBC__ == 2 && __GLIBC_MINOR__ > 2) 275 __c_locale __old = __uselocale(_M_c_locale_ctype); 276 #endif 277 wint_t __i; 278 for (__i = 0; __i < 128; ++__i) 279 { 280 const int __c = wctob(__i); 281 if (__c == EOF) 282 break; 283 else 284 _M_narrow[__i] = static_cast<char>(__c); 285 } 286 if (__i == 128) 287 _M_narrow_ok = true; 288 else 289 _M_narrow_ok = false; 290 for (size_t __j = 0; 291 __j < sizeof(_M_widen) / sizeof(wint_t); ++__j) 292 _M_widen[__j] = btowc(__j); 293 294 for (size_t __k = 0; __k <= 11; ++__k) 295 { 296 _M_bit[__k] = static_cast<mask>(_ISbit(__k)); 297 _M_wmask[__k] = _M_convert_to_wmask(_M_bit[__k]); 298 } 299 #if __GLIBC__ > 2 || (__GLIBC__ == 2 && __GLIBC_MINOR__ > 2) 300 __uselocale(__old); 301 #endif 302 } 303 #endif // _GLIBCXX_USE_WCHAR_T 304 305 _GLIBCXX_END_NAMESPACE_VERSION 306 } // namespace 307